Bibliographic citations
Bozzo, M., (2019). Visión artificial basada en redes neuronales convolucionales para detectar defectos en botellas vacías de vidrio para una planta embotelladora [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/20660
Bozzo, M., Visión artificial basada en redes neuronales convolucionales para detectar defectos en botellas vacías de vidrio para una planta embotelladora [Tesis]. PE: Universidad Nacional de Ingeniería; 2019. http://hdl.handle.net/20.500.14076/20660
@misc{sunedu/3926062,
title = "Visión artificial basada en redes neuronales convolucionales para detectar defectos en botellas vacías de vidrio para una planta embotelladora",
author = "Bozzo Cornejo, Manuel Ernesto",
publisher = "Universidad Nacional de Ingeniería",
year = "2019"
}
This present thesis “Artificial vision based on convolutional neural networks to detect defects in empty glass bottles for a bottling plant” is proposed to demonstrate, in a preexperimental way in an inspection line for a bottling plant, that can to detect the defects in empty glass bottles, using artificial vision based on convolutional neural networks. In a bottling machine, there is the inspection production line of the returnable empty glass bottles, before the final product to be packaged enters, this line is after the cleaning process (washing and disinfection) of these empty glass bottles; within the inspection, there are some defects, the two main ones are: a. Presence of persistent dirt stuck in the glass bottles. b. Presence of cracks in the glass bottles. To achieve the detection of these types of defects, is proposed in this thesis, the use of artificial vision by computer designed with a convolutional neural network, developed in the open source Python programming language.
This item is licensed under a Creative Commons License